m - nhanes_1988_2018.R

#############################################################################################################
###############################  MAIN SCRIPT - FORMING NHANES CURATED DATASETS  #############################
#############################################################################################################

#~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~#
#~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~  Assign Directories of NHANES Datasets  ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~#
#~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~#

mortality_directory <- paste(working_directory
                             , "Mortality Datasets"
                             , sep = "/") 

#~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~#
#~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~  Compile the NHANES Datasets  ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~#
#~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~#

# Upload the cleaning documentation for all datasets


# Extract the individual mortality datasets and compile them into the unclean mortality dataset
mortality_unclean <- compile_mortality_dataset(dataset_directory = mortality_directory
                                               , current_directory = working_directory)

setwd(working_directory)
response_unclean <- compile_datasets(cleaning_documentation = list_master_files$Response
                                     , current_directory = working_directory
                                     , name_dataset = "Response")

dietary_unclean <- NHANES_dietary_final

# response_unclean <- clean_duplicates_of_seqn_from_cycles(response_unclean)


demographics_unclean <- compile_datasets(cleaning_documentation = list_master_files$Demographics
                                         , current_directory = working_directory
                                         , name_dataset = "Demographics")

# demographics_unclean <- dumb_compile_demographics_datasets(list_document_cleaning = list_master_files
#                                                            , current_directory = working_directory
#                                                            , name_dataset = "Demographics")


# Fix this due to change from df to list for the cleaning documentation
medications_unclean <- compile_datasets(cleaning_documentation = list_master_files$Questionnaire %>%
                                          filter(grepl("\\bPrescription Medications\\b"
                                                       , .$file_summary) == TRUE &
                                                   is.na(.$SDDSRVYR) == FALSE)
                                        , current_directory = working_directory)


chemicals_unclean <- compile_datasets(cleaning_documentation = list_master_files$Chemicals
                                      , current_directory = working_directory
                                      , name_dataset = "Chemicals")


weights_unclean <- compile_datasets(cleaning_documentation = list_master_files$Weights
                                    , current_directory = working_directory
                                    , name_dataset = "Weights")

occupations_unclean <- compile_datasets(cleaning_documentation = list_master_files$Occupation
                                        , current_directory = working_directory
                                        , name_dataset = "Occupation")

setwd(working_directory)
questionnaire_unclean <- compile_datasets(cleaning_documentation = list_master_files$Questionnaire %>%
                                            filter(grepl("\\bPrescription Medications\\b"
                                                         , .$file_summary) == FALSE &
                                                     is.na(.$SDDSRVYR) == FALSE)
                                          , current_directory = working_directory
                                          , name_dataset = "Questionnaire")

#~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~#
#~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~  Clean the NHANES Datasets  ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~#
#~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~#

# Clean the mortality dataset
mortality_clean <- clean_mortality_dataset(mortality_unclean
                                           , list_master_files
                                           , "Mortality")

dietary_clean <- clean_dietary_dataset(dietary_unclean
                                       , list_master_files
                                       , "Dietary")

demographics_clean <- clean_demographics_dataset(demographics_unclean
                                                 , list_master_files
                                                 , "Demographics")

setwd(working_directory)
response_clean_test <- clean_response_dataset(response_unclean
                                              , list_master_files
                                              , "Response"
                                              , demographics_clean)

medications_clean <- clean_medications_dataset(medications_unclean
                                               , list_master_files
                                               , "Questionnaire")



chemicals_clean_test <- clean_chemicals_dataset(chemicals_unclean
                                                , list_master_files
                                                , "Chemicals")

list_chemicals_clean <- clean_chemicals_dataset(chemicals_unclean
                                                , list_master_files
                                                , "Chemicals")

chemicals_clean <- list_chemicals_clean$only_harmonized_variable

comments_unclean <- list_chemicals_clean$harmonized_and_unharmonized_variables

comments_clean <- form_comments_dataset(comments_unclean
                                        , list_master_files
                                        , "Chemicals")

weights_clean <- form_survey_weights_dataset(weights_unclean
                                             , list_master_files
                                             , "Weights")

occupation_clean <- clean_occupation_dataset(occupations_unclean
                                             , list_master_files)



# process_fix_categories(list_master_files$`Questionnaire Fix Category`)
# 
# check_num_cycles_documentation(list_master_files
#                                , "Questionnaire")

# process_lower_cases_fix_cats(list_master_files$`Questionnaire Fix Category`)

# check_new_codename(list_master_files
#                    , "Questionnaire")
# 
# eliminate_fix_categories_same(list_master_files$`Questionnaire Fix Category`)


questionnaire_clean <- clean_questionnaire_dataset(questionnaire_unclean
                                                   , list_master_files
                                                   , "Questionnaire")


#~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~#
#~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~  Create dictionary  ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~#
#~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~#

# List of tables on inconsistencies for each type of module or dataset
list_master_files <- upload_nhanes_master_files("NHANES - Master List of Files 1i.xlsx")

# Dictionary of drugs and their descriptors
df_medications_drug_info <- create_dictionary_drugs("RXQ_DRUG")

# Dictionary of all variables in curated NHANES datasets and their descriptors
df_dictionary <- create_dictionary(list_dataset = list("mortality" = mortality_clean
                                                       , "dietary" = dietary_clean
                                                       , "demographics" = demographics_clean
                                                       , "response" = response_clean
                                                       , "medications" = medications_clean
                                                       , "questionnaire" = questionnaire_clean
                                                       , "chemicals" = chemicals_clean
                                                       , "occupation" = occupation_clean
                                                       , "weights" = weights_clean
                                                       , "comments" = comments_clean)
                                   , list_documentations = list_master_files)

# Dataset of harmonized levels for categorical variables
df_levels_categorical_variables <- create_dictionary_harmonized_categories(list_documentations = list_master_files)